EVALUATION OF THE UTILIZATION OF THE NATIONAL CASH REGISTER PERSONNEL TEST AND THE PICTORIAL REASONING TEST IN THE EMPLOYMENT SELECTION PROCESS IN COLONIAL STORES, INCORPORATED BY ILIAD L. CHANAY, JR. A professional paper submitted to the Graduate Faculty in partial fulfillment of the requirements for the degree of MASTER OF EDUCATION MONTANA STATE UNIVERSITY Bozeman, Montana August, 1968 IHESES P31S ACKNOWLEDGEMENTS . . . to every man his chance — to every man, regardless of his birth, his shining, golden opportunity — to every man the right to live, to work, to be himself, and to become whatever thing his manhood and his vision can combine to make him -- this . . . is the promise of America. - Thomas Wolfe The writer gratefully acknowledges the invaluable assis¬ tance and encouragement provided by three members of the University faculty: Dr. Kenneth J. Goering, Dean, College of Graduate Studies; Dr. E. N. Ringo, Dean, School of Education; Dr. M. S. Monson, Professor, School of Education. Appreciation is also due colleagues at Colonial Stores, Incorporated, and others interested in the subject matter of this paper who helped the writer both by their expressions of encouragement and helpful suggestions. m TABLE OF CONTENTS Page VITA ii ACKNOWLEDGEMENTS iii LIST OF TABLES vi LIST OF FIGURES vii CHAPTER I. INTRODUCTION 1 Statement of the Problem 1 Procedure 1 Scope and Limitations 5 Definition of Terms 6 II. BACKGROUND INFORMATION ON COLONIAL STORES, INC. 7 History of the Company 7 Operations 7 Management 8 Employee Relations 9 Improvement and Modernization Plan 9 Sales, Earnings, and Dividends 10 Employees 10 III. THE EMPLOYMENT PROCESS 11 Recruiting 11 Interviewing 13 Applications and References 14 Testing 16 Personnel Decisions 21 IV. DESCRIPTION OF TESTS 25 National Cash Register Personnel Test ... 25 Science Research Associates Pictorial Reasoning Test 30 iv TABLE OF CONTENTS - Continued. CHAPTER Page V. ANALYSIS AND INTERPRETATION OF DATA 42 Availability of Test Results 43 Description of the Sample Groups 45 Comparison of Scores on Personnel Test and Pictorial Reasoning Test 46 Employee Ratings on NCR Battery, Pictorial Reasoning Test 49 Employee Ratings on the National Cash Register Battery Test, Pictorial Reason¬ ing Test, and Personnel Rating Sheet. ... 49 VI. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ... 53 Summary 53 Conclusions 55 Recommendations 56 APPENDICES 61 LITERATURE CITED 62 v LIST OF TABLES TABLE Page 1. Pictorial Reasoning Test-Summary Data 33 2. Comparing Means of Urban-Nonurban and White- Negro in Untimed Administration Pictorial Reasoning Test 35 3. 1967 Untimed Norms for the Pictorial Reasoning Test 39 4. 1967 Timed Norms for the Pictorial Reasoning Test 40 5. Applicants Scoring Less Than 400 on the National Cash Register Test Battery, Raw Score and Percentile Ranking on the Pictorial Reason¬ ing Test 47 6. Applicants Scoring 401 and Above on the National Cash Register Test Battery, Raw Score and Percentile Ranking on the Pictorial Reasoning Test 48 7. Employees' Ratings on the National Cash Register Test Battery, Pictorial Reasoning Test and Personnel Rating Sheet 49 8. Supervisor's Rating of Employee's Job Performance 51 vi LIST OF FIGURES FIGURE Page 1. A Diagram of Steps Taken in a Typical Employ¬ ment Procedure 12 vn CHAPTER I INTRODUCTION Statement of the Problem We are living in an age of tests. Almost as soon as one enters kindergarten the testing process begins. Even before we can read we are subjected to different types of picture tests to determine the intelligence level to which we are assigned. Throughout our life we take test after test. Many jobs are filled by aptitude and pre-employment tests, and in many areas promotion is determined largely by the ability to pass tests. Before a young man is drafted, he is tested and re-tested. To drive an automobile we must pass a performance test. Many persons today are successful simply because they have become expert "test-takers." In recent years, increasing attention has been focused on testing in the personnel selection process in today's companies. The term personnel selection techniques implies the idea of choice — that there is a group of persons (appli¬ cants) and a group of jobs (either entry jobs or promotions) that these applicants are seeking, and that personnel men are in a position to choose among the applicants to fill the jobs. We assume, although in today's labor market it may not necessar¬ ily be the case, that there are more applicants than jobs. Confronted with a pool of applicants and a list of jobs, 2 the personnel man must decide on some method of selection. He could hire or promote people in the order of their appear¬ ance, in alphabetical order, or in seniority order. Hiring people in order of their appearance or in alphabetical order would be almost purely random, with no attempt to determine who is better qualified or more likely to succeed. Of course, it is nondiscriminatory, in all senses of the term, but it can hardly be called selection. Another approach is to select on the basis of some kind of prior personal contact — employee referral, nepotism, the network of friends and acquaintances that exist in any community. Or the personnel man may rely heavily on reference checks and on previously determined standards, such as a high school diploma, or maximum or a minimum age, or some other rigidly defined and applied criterion or combination of criteria. Then there is the use of the personnel interview, based on the assumption that a personal assessment by a skilled interviewer can add something to the chances of securing the best employees from the available pool of applicants — some¬ thing beyond what can be accomplished through standards, reference checks, or referrals, for instance. There is also on-the-job trial, or a probationary period; this may be the best possible selection method, but neither 3 employers nor applicants seem especially enthusiastic about it, preferring to assume that once someone is hired, the job is his. In addition to these selection methods, there is psy¬ chological testing. Testing alone is probably as effective as any of the others alone and may be better, and testing in conjunction with standards and interviews and reference checks probably increases the chances of a successful hire consider¬ ably. It should be emphasized, however, that testing alone as a criteria for selection is not a defensible method. The objective of personnel men should be to develop a selection system that, in each of its steps from recruting through placement and training, does the best possible job of finding, hiring, training, and placing persons in accordance with their skills and abilities and without regard for personal characteristics such as sex, race, color, religion, national origin, or, in some states, age (which now, by law, may not be considered). The purpose of this paper was to describe the efforts of one corporation to re-evaluate its personnel selection procedures, specifically the use of tests as screening devices, in light of changing needs and conditions of employment. The purpose of the study was to evaluate the effectiveness of the National Cash Register Personnel Test as a predictive 4 measure of probable job performance; general intelligence and ability to learn on the job; employee's potential for advance¬ ment, and further to answer the following questions: Did the test give predictions of performance for various job classifications more often than not? Did the test predict accurately for all groups in the population, including minor¬ ity group members? Procedure In researching this paper, library materials, such as books, pamphlets, reports, periodicals, newspapers, and magazine articles were consulted extensively. Interviews were also held with various professional personnel involved in employee selection. The National Cash Register Personnel Test was administered to a group of thirty persons, randomly selected, from appli¬ cants during the month of November, 1967, to the Atlanta division. Additionally the Pictorial Reasoning Test was administered to the same applicants in the following groupings: Group A - Applicants scoring 400 or less on the Personnel Test; Group B - Those applicants scoring 401 and above. The two tests were also administered to a group of thirty- five (35) employees, randomly selected from the Atlanta Division of Colonial Stores. In this group, a Personnel Rating Sheet, 5 constructed by the writer was completed on each person in the sample by his or her supervisor. A copy of the Person¬ nel Rating Sheet is included in the appendix. This latter procedure was followed to determine the possible relationship between ability as measured by the Personnel Test, the Pictorial Reasoning Test, and on-the-job performance. Appro¬ priate tables have been constructed in an effort to record the information on test results in a uniform manner. Scope and Limitations This study was concerned with thirty applicants for employment at Colonial Stores, Inc., as well as thirty-five employees. This was an evaluative study in which the writer attempted to point out through analysis of test results, the effectiveness of the National Cash Register Personnel Test and Science Research Associated Pictorial Reasoning Test. In all subsequent discussions, these tests shall be referred to as NCR and PRT, respectively. The sample for the conclusions drawn in this study was limited to the applicants and employees designated above. Due to the limited time allotted for the study, it was impossible and impractical to attempt an in-depth factoring of the data. Available library materials were limited. The study was further limited by the inexperience of the writer, both in the methods of research and in the employee selection process. 6 Definition of Terms For the purpose of this investigation, terms used are defined as follows: 1. Percentile — a value on a scoring scale at or below which a given percentage of cases fall. 2. Timed test — a test for which there is a required specific time limit for each sub¬ section . 3. Untimed test — a test which has no speci- time limit either for the various sub-sections or the entire test. CHAPTER II BACKGROUND INFORMATION HISTORY Colonial Stores was incorporated as the David Pender Grovery Company in Virginia on January 24, 1901. The Colonial name was adopted December 19, 1940, when the David Pender Grocery Company, with principal operations in Virginia and North Carolina, was merged with the Southern Grocery Company, operating principally in Georgia and South Carolina. The latter company had been incorporated in Delaware in 1925, and had acquired the business and properties of the L. W. Rogers Company, founded in Atlanta in 1892, and certain other chain stores. Through merger on June 18, 1955, with Albers Super Markets, incorporated in Ohio in 1933, Colonial extended its operations into the midwest. OPERATIONS Colonial Stores currently operates 414 supermarkets or discount grocery stores, serving over two million customers weekly from locations in the following nine states: Georgia North Carolina Virginia Ohio South Carolina Florida Kentudky Alabama Maryland 8 In addition/ through a wholly-owned subsidiary/ Galaxy Drugs, Inc./ the company operates thirteen drug stores — nine in the Atlanta area? one in Richmond, Virginia; one in Newport News, Virginia; one in Colonial Heights, Virginia; and one in Rockingham, North Carolina. All of the company's supermarkets are of the conventional self-service, cash-and-carry type, with the exception of 26 stores which are operated as discount grocery stores. Most of the supermarkets in Ohio and Kentucky operate under the Albers name. The Colonial name is used in other states. Some of the discount stores are identified as Big Star Stores. Colonial has six operating divisions, each under a divi¬ sion vice president responsible for from 34 to 79 stores. Division headquarters are located in Atlanta, Georgia; Jackson¬ ville, Florida; Columbia, South Carolina; Raleigh, North Carolina; Norfolk, Virginia; and in Cincinnati, Ohio. Although a Virginia corporation, the company's corporate headquarters are located in Atlanta. MANAGEMENT Ernest F. Boyce, an executive with over thirty years experience in the retail food and drug business, is president of the company. Corporate headquarters responsibilities are divided between vice presidents for merchandising, operations, stores 9 development and properties, finance, manufacturing, ware¬ housing, and transportation and public relations. Division vice presidents have a high degree of autonomy. They report directly to the corporate vice president-operations. Each division is an operating entity with the exception of accounting which is done on a regional basis. EMPLOYEE RELATIONS The company and its wholly-owned subsidiaries employs approximately 12,506 full and part-time employees. All hourly employees in the company's distribution centers and bakeries, as well as the majority of hourly employees in the retail stores, are represented by labor unions with which the com¬ pany has maintained favorable relations. The company negotiates individual union contracts with labor union locals in the various areas in which it operates. There have been no mate¬ rial work stoppages due to labor disagreements in recent years. Management of personnel participates in a Bonus Incentive Plan. The company maintains an active personnel recruiting and training program. IMPROVEMENT AND MODERNIZATION PROGRAM During 1966, Colonial opened twenty-two new supermarkets and, under the Galaxy name, one retail drug store. Forty-two supermarkets were enlarged or remodeled, and twenty-eight older 10 and less productive units were closed. Fourteen new supermarkets were opened in 1967; twenty-two closed or sold. In 1968, twenty-two new retail food stores are planned. SALES, EARNINGS, AND DIVIDENDS Colonial has continuously paid dividends since 1940. The current annual rate is $1.40 on common and $2.00 on preferred. Colonial announces sales and earnings at the end of each six months. Dividends have been paid quarterly. EMPLOYEES Colonial Stores, Inc., had a total employment force of 12,506 in 1967, 9,228 males and 3,278 females. Of this number, 1,252 were Negro; 1,086 males and 166 females. CHAPTER III THE EMPLOYMENT PROCESS Testing is only one small part of the total employment process, although there are important interrelationships between testings and other aspects. A typical pattern of employment activity is diagrammed in Table 1 on the following page. Although details of the procedure may vary from one organization to another, the pattern is typical. Perhaps the most important feature of the process is that the decision rests essentially upon rejection; if an applicant is not rejected at some point in the succession of hurdles, then he is hired. Recruiting.—Recruiting gets applicants into the office. Some recruiting policies include (1) promoting or transferring employees from within; (2) making a special effort to hire the handicapped or members of minority groups; and (3) adver¬ tising or listing job opportunities with specialized employment agencies. Less obviously but equally important, recruitment is based on company reputation and other less tangible forces within the community. Stromberg found, for example, that one result of a testing program is an improvement in the caliber of applicants coming to the employment office. He suggested that '^'Stromberg, E. L. "Testing Programs Draw Better Appli¬ cants," Personnel Psychology, I, pp. 21-29. 12 FIGURE I A Diagram of Steps Taken in a Typical Employment Procedure 1 1 Uhrbroch, R. S. "Mental Alertness Tests as Aids in Selecting Employees," Personnel, XII, pp. 229-237. 13 the use of tests establishes, as part of the company reputa¬ tion within the community, the idea that the company seeks and is getting the best people in the labor force. Recruiting also includes the simple device of keeping the door to the employment office open so that applicants may walk in off the street. The characteristics of the imme¬ diate neighborhood do much to determine the kinds of applicants this "recruiting" brings in. Changes in the community (e.g., racial or national composition, socio-economic status) will be reflected by changes in the applicants coming in for employ¬ ment . Recruitment is important to personnel testing since its procedures, explicit or implicit, define the nature of the applicant population upon which research is done. Changes in recruitment policy and activity result in changes in the applicant population. A test that is valid on a sample of the kind of applicant population recruited at one time may prove to be less valid, or perhaps still more valid, as the applicant population changes. Interviewing.—Interviews and interviewing procedures are notoriously fallible. Even so, the interview is the focal point of the employment process. No part of the process, in¬ cluding testing, is fool proof. The interview is needed as a place where subjectivity may deliberately enter the process 14 before decisions are reached. Interviewing provides for a preliminary screening of applicants. The obviously unfit applicant should not be encouraged to waste his time/ nor should the company want to expend its time and resources in false encouragement. The preliminary interview may result in the immediate rejec¬ tion of the few obviously unqualified applicants, and for the others, it can serve as an orientation tool by which direction is given to the rest of the employment process. If tests follow a screening interview, then the interview, like recruiting, defines the population to be studied by the test administrator. The influence of the preliminary interview on the client's subsequent performance can be crucial. Depending on the con¬ duct of the interviewer, an applicant can become tense and frightened or he can become relaxed and confident. He may become more motivated to get the job, or he may approach the tests with a defeatist attitude. Sound personnel policy dictates that the interviewer try to increase motivation and to eliminate unnecessary stress and fear. The influence of the interview should be kept as constant as possible for all applicants. Applications and References.—In nearly all organizations some sort of application form filled out by the applicant is 15 used to help in assessing the applicant's background and qualifications. The interviewer examines the completed form and makes certain judgments which are seldom very precise. They are more likely to be simple generalities, such as "this is a good guy," or "this one doesn't look so hot." At a higher level of sophistication, put to an empirical test. An individual's background can also be assessed by check¬ ing his references. A telephone conversation will often get information far more reliable and accurate than will be ob¬ tained in writing, but even written references can be useful if the writer is asked to answer specific questions and is assured that his replies will be confidential. The typical letter of recommendation written upon request of the applicant is often useless. It is likely to be filled with glittering generalities from which no reliable references may be drawn. When such a letter is negative, of course, it can be useful simply because this is so rare. The local credit bureau is a reference source that is frequently overlooked. Credit bureaus usually have rather full information on any applicant who has been known to the community for any length of time. An applicant who appears promising in the employment office seem less desirable when it is learned that he habitually neglects financial responsibili¬ ties . 16 These references, of course, do not help to choose between otherwise equally good applicants as much as they help in finding out who the most undesirable applicants are and in getting rid of them. Rejection is not always based upon predicted job performance; it may also be based upon policy. An applicant, for instance, who is considered immoral or one who is physically dirty may rank high in terms of predicted performance ability, yet be rejected on the basis of a company policy against having immoral or unclean people in the organization. Testing.—Tests are simply a part of the process. They can be a big help when the final personnel decision is reached, but they do not make the decision. An employment manager who relies blindly on tests, who will not under any circum¬ stances temper test results with his own judgment, has abdicated his responsibility. This is not to suggest that the tests have no value; in fact, where test scores and other quantitative data strongly suggest a particular decision, it will rarely be wise to go against that decision. Nevertheless, rare events do occur, and the final responsibility for deciding whether a particular case represents the usual thing or the rare event rests with the employment manager, not with the constructor. The point in the selection process at which to test varies from one company to another; there seems to be little 17 objective evidence to indicate the tests are either more or less effective when they are administered early or late. There is, however, one aspect of timing which is most impor¬ tant: if a trial test battery— the collection of tests that have not yet been validated and are not being used for actual selection — is administered at a particular point in the process then the battery finally adopted for actual selection should be used at that same point. The reason refers again to the sampling of the applicant population. The applicant population at the beginning of the process is a relatively unselected one. At each successive stage, however, there are vague and perhaps poorly defined changes in the nature and homogeneity of that population which may make significant changes in the patterns of test results.'*' Personnel testing has one specific objective: to con¬ tribute to the increasingly effective use of manpower within an organization. At all manpower levels, the effectiveness of utilization must be guaged by one or both of two kinds of measures: (1) how will people do on their jobs; and (2) how much personal satisfaction does each individual find in the ^"Kellogg, Marion S. , Closing the Performance Gap (New York: American Management Association, 1967), Chapter II. 18 performance of his job. These two measures are quite general. They suggest many more specific questions to which, through testing, answers can be predicted for individual applicants: How fast will he work? How much skill will he develop through training programs? How often will his work be marred by errors? How consistent will his production be? What kinds of attitudes will he develop toward his work? How well will he be motivated to produce? How long will he stay on the job? How regular will he be in attendance? By permitting predic¬ tion to answers to such questions as these, tests can contri¬ bute significantly to the improved selection and placement of applicants. It should be made quite clear at the outset that person¬ nel testing does not provide completely accurate prediction of performance on the job. In the first place, variations from one worker to another are due to many causes. These may include attitudes toward supervisors, working conditions, or training, as well as the possession of required aptitudes and motives at the time of hiring. There are many other personnel activities besides selection and placement that try to improve the use of manpower: industrial engineering; wage and salary '*'Katzell, R. A. "Industrial Psychology," of Psychology, VIII, 237-268. Annua1 Review 19 administration; or attitude surveys, suggestion systems, and other techniques for taking the industrial relations pulse. Beyond these necessary limitations, accuracy is further limited by the fact that prediction must be based upon im¬ perfect measurement. Measures of satisfaction, of production, of aptitudes, and of motives — all such measures are subject to considerable error. In short, it is unrealistic to expect any selection program to be perfect. Nevertheless, it is both realistic and essential that efforts continually be made to improve employment prediction. Prediction of future behavior, whether mathematical or merely implicit and subjective, forms the basis on which an employment interviewer or supervisor makes a decision about an individual applicant: to reject him, to hire him, to reject him for one job but consider him further for another, or to place his name on a waiting list. Psychological testing is finding ever-increasing use in American business, competent use of tests and test techniques applied to other parts of the employment process can help industry take long strides toward the effective utilization of manpower. In some instances, however, psychological testing is finding ever-increasing abuse in American business. Testing programs are installed merely because they are stylish. Tests are chosen because of catchy names or clever promotion, without considering what they might be measuring, if anything. Tests 20 are given, and applicants are selected, without any effort being made to discover the facts about what the tests will do. The real experts, however, are very much concerned with find¬ ing out for sure whether a test will help make good predic¬ tions of on-the-job behavior. Several important factors must be taken into consideration when selecting tests: 1. Validity or "testing the test."'*' Does the test measure what it purports to measure? No test does this perfectly for an indivi¬ dual since it is, after all, a sampling technique based on group data, but the odds should be in favor of its doing what it says it does. 2. Reliability. If a person were to take the same tests again, memory factors aside, is he likely to get the same score? If the answeris a negative one, the test results are obviously meaningless and cannot be used. 3. The norm group. That is, against the per¬ formance of what kind of people was the test validated? If it was a test of creativity and was validated against the scores of a group of accountants, it is unlikely to be useful in predicting the creativity of engineers.2 3 The "mental measurement duffers" are not concerned with these factors. They pick tests on theory or on looks or be- '*'Guion, Robert M. , Personal Testing (New York: McGraw- Hill Book Company), p. 3. 2 Kellogg, Marion S., op cit., pp. 94-95. 3 Guion, Robert M., op. cit., p. 5. 21 cause of someone's testimonial and are satisfied as soon as the choices are made. Since most tests are fairly good in some respects/ and since almost anything is a little bit related to almost everything else, the tests chosen in this way sometimes really do provide a better work force. Most of the time they have no relationship either to performance or to satisfaction; they do no particular good. With appalling frequency, however, the incompetent testing program actually results in the selection of applicants who are less satis¬ factory . Personnel Decisions.—A decision to hire, not to hire, or to hold on a waiting list is the end product of the employ¬ ment process. Cronbach and Gleser show clearly that this is the basic problem in employment and the problem on which test¬ ing programs and future research must focus.^ They point out that the principle stated by Hull that, "the ultimate purpose of using aptitude tests is to estimate or forecast 2 aptitudes ..." has been the foundation of nearly all work on test theory. Cronbach and Gleser prefer to abandon this Cronbach, L. J., and Gleser, Goldine C., Psychological Tests and Personnel Decisions (Urbana: University of Illinois Press), p. 157. 2 Hull, C. L., Aptitude Testing (New York: Harcourt, Brace and World, 1928), p. 206. 22 point of view. They accept the usefulness of accurate esti¬ mation, or prediction, but they insist that the ultimate purpose of testing, and of the employment process generally, is simply to arrive at a decision to hire or not to hire. This evaluation takes a special significance in view of the guidelines for employment testing procedures recently set forth by the Equal Employment Opportunity Commission. The Commission: . . . encourages the use of job-related ability tests. Employers must be aware that when an applicant has not enjoyed equal educational and developmental oppor¬ tunities, his score on a test may under¬ estimate his job potential. The ultimate standard, however, is not the test score, but performance on the job. Since cultural factors can so readily affect performance on so many tests, it is recommended that tests be judged against job performance, rather than by what they claim to measure. The sample population (NORMS) used in validating the tests should include repre¬ sentative members of the minority groups to which the tests will be applied. Only a test which has been validated for minori¬ ties can be assumed to be free of inadvertent bias. •*- The point of these experts is well taken; there is certain¬ ly a real need for more research on applying decision theories to personnel testing. Nevertheless, this view needs to be tempered somewhat. Although there is seldom any interest in 1Equal Employment Opportunity Commission, Guidelines on Employment Testing Procedures (Washington, D. C.: EEOC, August 24, 1966), p. 4. 23 literal predictions of future performance# the personnel decision to hire or not to hire is based upon at least an implicit prediction of the behavior lively to be shown on the job, A technically superior testing program can fail. Test¬ ing must fit into the particular administrative climate of the organization in which it is applied. Development of a testing program must therefore follow many steps. Some of these are technical, others are administrative. Basically, the steps can be listed as followst 1, Establish policies and get support from management, supervision, and labor union, 2, Put the program in charge of a qualified person, 3, Identify needs, 4, Select a trial test battery, 5, Validate the tests, 6, Incorporate the procedure into the employment process. 7, Make provision for periodic review,* There is no question that testing is a valuable selection instrument which — when properly used — can provide an *Thi8 list follows, with some modification, the views of Lawshe, C. H,, Principles of Personnel Testing (New Yorks McGraw Hill, 1946). In his presentation, he made a stronj plea for competence throughout, he pointed out that it is not enough to be "merely interested in testing," 23 literal predictions of future performance, the personnel decision to hire or not to hire is based upon at least an implicit prediction of the behavior likely to be shown on the job. A technically superior testing program can fail. Test¬ ing must fit into the particular administrative climate of the organization in which it is applied. Development of a testing program must therefore follow many steps. Some of these are technical, others are administrative. Basically, the steps can be listed as follows: 1. Establish policies and get support from management, supervision, and labor union. 2. Put the program in charge of a qualified person. 3. Identify needs. 4. Select a trial test battery. 5. Validate the tests. 6. Incorporate the procedure into the employment process. 7. Make provision for periodic review.'*' There is no question that testing is a valuable selection instrument which — when properly used — can provide an 1This list follows, with some modification, the views of Lawshe, C. H., Principles of Personnel Testing (New York: MeGraw Hill, 1948). In his presentation, he made a strong plea for competence throughout, he pointed out that it is not enough to be "merely interested in testing." 24 employer with objective information on the abilities of job applicants and increase the likelihood that those selected will perform effectively on the job. Tests can also benefit prospective employees by guiding them into the right jobs for their abilities and interests. A research study by the United States Equal Employment Opportunity Commission asserts that: . . . careful selection and administering of tests and validation of the testing instrument within an industrial setting may be the most desirable means to achieve the goal of full utilization of the nation's human resources.^ Phillip Ash, research assistant to the vice president for industrial and public relations of Inland Steel Company summarized: The time has come to remedy the all too-common practice of taking a brief intelligence test off the shelf and using it for all jobs, without local norms, without criterion data relationships, and, in fact, frequently with¬ out any demonstrable relevance to the selection problem at hand.^ 1Equal Employment Opportunity Commission, op. cit, p. 5. 2 Ash, Phillip, "Selective Techniques and the Law," Personnel, XLIV (November-December, 1967), p. 11. CHAPTER IV DESCRIPTION OF TESTS The National Cash Register Check-Out Test Battery In April, 1959, the National Cash Register Company made a research grant to the management consulting firm of Ward J. Jenssen, Incorporated, for the purpose of developing a battery of tests to be used as an aid in the selection of check stand operators. After nearly two years of research, the Check-Out Cashier Test Battery was developed. The Battery consists of two major sections: The Personnel Test for Prospective Check-Out Cashiers and the Personal Opinion Ballot. The following brief review of the contents of each of the two major sections reveal what information the test provides concerning the testee. The Personnel Test for Prospective Check-Out Cashiers This part of the Battery is an aptitude test containing three separately timed subtests. A copy is included in the appendix. Part 1: The Register Tape Checking Test.—This subtest measures ability to quickly perceive minor details; to recog¬ nize subtle differences in similar appearing material. This test, according to the authors, correlates well with both accuracy of work and speed of performance as a check-out cashier. 26 Part 2: The Non-Verbal Reasoning Test.—This subtest measures ability to deal with new and unusual problems; to quickly adapt one's thinking to different kinds of situations. The test does not measure the kind of "intelligence" which is acquired through education. Rather, it evaluates the applicants ability to think clearly and to handle practical, everyday matters, regardless of the degree of formal education to which the applicant has been exposed. Part 3: The Register Key X-ing Test.—This subtest measures both eye-hand coordination and skill at quickly performing a simple perceptual test. Scores on this test provide a good measurement of a person's speed of performance as a check-out cashier. The Personal Opinion Ballot The Personal Opinion Ballot (not utilized by Colonial Stores in its selection process) enables the applicant to express his views regarding himself and his environment. It is unlikely that there could ever be agreement among employers as to what constitutes the theoretically "ideal" checker. It was found, however, that the check-out cashiers who received higher ratings in customer relations, supervisory relations, and relations with fellow employees tended to answer certain key questions on the Personal Opinion Ballot in one way, while checkers with lower ratings in these interpersonal relationships 27 answered in another way. Administration and Scoring Each testee should receive one test booklet and sharpened pencils with erasers.* The test should be given in a room free from distractions, with comfortable seating, adequate separation, if more than one person is being tested, and good lighting. Each section of the test is timed separately, and established times should be strictly followed. These standard conditions should be adhered to or the results will be invalid.** Instructions for scoring the Personnel Test and inter¬ pretation of the test are included in the Manual. Interpretation of Scores The raw scores from the National Cash Register Company Test Battery can be interpreted in one of two different ways as in the employment decision. Each of the two methods are described in detail in the Manual. Employment managers can decide which of the two methods is best suited to their own situation. Analysis of Subtests Using Percentiles.—A simple way to analyze the subtests is to translate raw scores into percentile *Each person should be asked to fill in the information required on the back cover. **Testee should be permitted to relax a few minutes follow¬ ing the completion of each section of the test. 28 or rank-standing scores. As an aid in interpreting the percentile scores, approximate categories have been indicated. These categories are merely suggestive. After gaining skill in working with the test, employment managers may wish to revise these suggested categories to better suit his own company's employment standards. Use of a Single Weighted Score.—The second way to use the NCR Test Battery is to combine all subtest scores into one composite score which can then be used in assessing the applicants's probable over-all job performance. The basic assumption underlying this method of interpretation is that if the job performance ratings in Accuracy, Speed, and Inter¬ personal Relations are totaled, the sum would represent a meaningful index of over-all job performance. If an employ¬ ment manager is willing to make that assumption, then the single weighted score can be used to assess an applicant's probable job performance. One common method of putting the final, weighted composite score to practical use is to employ it as a "cutting score." This simply means that any applicant who scores at or above a certain level will be hired; if they score below this critical or cutting score level, they will not be hired. This procedure presumes, of course, that the applicant satisfied other criteria of employment such as personal appearance, job references, 29 etc. Meaningful results from the use of this test can be obtained only if the employment manager sets his own cutting score — the score that will be best to use for his own organization under current conditions. In setting the cutting score, at least four relevant considerations should be evaluated: 1. How urgent a need is there for employees? Can the company afford to wait for the very "cream" of job applicants? 2. What general caliber of applicant does this company normally have coming in for a job? What can they expect in the way of a better applicant if they wait a little longer? How much does it cost to attract job appli¬ cants to the employment office? 3. What is the condition of the available labor market? In a labor market where skilled checkers are readily available, one can afford to reject many applicants, holding off until a high scorer comes along. In a labor market where good checkers are scarce, however, it quickly becomes uneconomical to maintain unduly high cutting score standards. 4. Does the company want to raise its personnel standards and, therefore, hire only appli- ^ cants who have the greatest chance of success? Jenssen, Ward J., NCR Test Battery Instruction Manual for Prospective Check-out Cashiers (Dayton, Ohio: The National Cash Register Company, 1961), pp. 10-11. 30 The Science Research Associates Pictorial Reasoning Test The SRA Pictorial Reasoning Test is a test of general ability that was empirically developed to be used with all major American subcultural groups. This eighty-item test can be used to measure the learning potential of individuals from diverse backgrounds with reading difficulties, whose potential for training and employment cannot be reliably and validly measured by verbal instruments. It was authored by Robert N. McMurry and Phyllis D. Arnold. The test has been designed for individuals fourteen years of age or older. It is applicable particularly in pre¬ dicting performance in situations that do not require previous¬ ly learned reading skills. Test Development Two hundred fifty-three pictorial items were developed. Each item consists of a series of five pictures, four of which are related by an underlying principle or theme. A fifth picture does not fit into the general theme and is therefore the correct choice as the answer. These items were designed to cover a wide range of difficulty. Sixty items constituting the current SRA Nonverbal Test, together with the 253 new items, were administered in an interview setting to between twenty and fifty members of seven 31 American subcultures. The subcultures interviewed included Appalachians, Spanish-speaking bilinguals, rural and urban Negroes, rural and urban whites, and French-speaking bilingual Canadians. The interviewing was done to determine which items contained ambiguities or unfamiliar material. During the interview the individual was asked to respond to each item, explain his reasoning, and comment on any picture with which he was unfamiliar. Pursuant to the results of these inter¬ views, the items were extensively edited and revised. Of the possible 313 items, 270 were selected for use in the three ninety-item experimental tests. The three experimental forms were administered in an untimed format to approximately six thousand members of the seven subcultures. As a reference standard for the more difficult items, five hundred college students also took the test. A point-biserial item analysis was performed separately for each subculture. The final items for the test were selected to maximize two criteria simultaneously. The first was the traditional criterion of choosing those items which best discriminate between the high and low scorers on the test; the second criterion was the requirement that each item have equal difficulty for each subculture. In this way the total 32 test was designed to be fair to all subcultural groups by selecting items that were individually, or in combination, comparable across cultures. The experimental pool consisted of nonverbal material that was as culturally unbiased as possible. Considering the subcultures' mean scores on all 270 items, it was found that the highest-scoring subculture (Control Group, 46.5) had a mean score of 31.4 per cent above the lowest-scoring subculture (Rural Negro, 35.4). After applying the criteria for item selection described above, the 80 items that now constitute the SRA Pictorial Reasoning Test had a mean for the highest group (Urban White, 53.1) that is only 8.7 per cent higher than that of the lowest group (Rural Negro, 48.9). These figures are based on the untimed experimental data. For normative purposes, approximately five thousand students, including dropouts and reform school students, took the PRT in timed and untimed situations. The range of school¬ ing completed was from grades eight through twelve; ages ranged from thirteen through nineteen. The group was composed of 50.1 per cent females and 49.9 per cent males, representing all seven subcultural groups. Cultural Fairness of the Pictorial Reasoning Test Table 1 summarizes the data obtained from the timed and untimed administrations of the PRT. In the timed condition, 33 all seven groups had nearly the same mean and distribution. As expected, the Control Group of nonurban whites (rural, suburban, and smaller cities) had the highest mean, and the Urban Whites had the second highest. TABLE I PICTORIAL REASONING TEST SUMMARY DATA Timed Administration Untimed Administration Mean Standard Deviation N Mean Standard Deviation N Control Group 47.7 6.5 1337 53.0 6.8 353 Urban White 46.0 7.3 318 55.6 6.7 198 Appalachian 43.7 8.9 147 53.3 6.6 376 Urban Negro 43.2 7.1 461 55.4 5.7 63 Rural Negro 41.9 8.5 309 47.3 9.4 201 Spanish 43.6 7.5 841 54.9 6.2 59 French 45.0 5.8 185 — — — TOTAL 45.2 7.4 3598 52.8 7.6 1250 In the second half of the summary table for untimed ad¬ ministration the groups (with one exception) are even closer in terms of means, and note also that the Control Group mean is exceeded by the means of four of the five other groups. Thus in this untimed condition the test appears to yield com- 34 parable results for the majority of the groups. In the untimed situation this relative improvement of the Other Groups is attributed primarily to the elimination of the Control Group's artificial increase in score associated with a general familiarity with tests, especially timed tests. The reason for the slightly lower mean of the Rural Negro sample can only be hypothesized. Perhaps this sample is biased, or perhaps there is a small but real difference because of selective migration. In Table 2, comparing the two factors of urban-nonurban and white-Negro in an untimed situation, the PRT shows greater mean difference between urban versus nonurban than Negro versus white. Thus, based on this data, urban groups regardless of race seem to have a better advantage than nonurban groups on these test items. It would appear that the subcultural fairness of the PRT is maximized by administering the test in an untimed condition. Administering the Test The SRA Pictorial Reasoning Test is self-administering. Each person who takes the test should be given ample time to read the general directions and sample problems and to work the practice problems before the testing session begins. The ex¬ aminer is to read all the instructions aloud while the examinees read them silently. These directions have been written at the fourth-grade reading level to facilitate understanding. TABLE II COMPARING MEANS OF URBAN-NONURBAN AND WHITE-NEGRO IN UNTIMED ADMINISTRATION Urban Nonurban Urban White 55.6 Nonurban White 53.0 Urban Negro 55.4 Nonurban Negro 47.3 Average 55.5 Average Difference: 5.4 50.1 White Negro Urban White 55.6 Urban Negro 55.4 Nonurban White 53.0 Nonurban Negro 47.3 Average 54.3 Average Difference: 3.0 51.3 Each examinee should have a test booklet and two lead pen¬ cils. Before the directions are read with the examinee, he should print his name on the front cover of the text booklet. The examiner is to read the indented sections aloud. Scoring the Test The SRA Pictorial Reasoning Test is scored by counting the responses automatically recorded on carbon as correct on the two 36 scoring grids. This recording takes place as the examinee marks his responses to the problems. Tear open the perforated edges on either side of the cardboard insert. Discard the carbon and begin scoring pro¬ blems 1-40. The score is obtained by following the chain of scores and counting the number of X's within the boxes. X's outside the boxes or X's with a circle drawn around them are not to be counted. Record this score in the box labeled Number Correct 1-40. Carry this score over to the second scoring grid page and write it in the first box located in the lower right corner. Follow the same procedure for scoring the second 40 problems. Add the two scores to obtain the total number correct, which is recorded in the third box. This total score can be converted to a percentile score by using the percentile tables. The test score is converted to a percentile to aid in interpretation. Interpretation Interpretation of Norms.—Test scores are most frequently interpreted interms of percentiles. A percentile is a value on the scoring scale at or below which a given percentage of cases fall. For example, the eightieth percentile is the point at or below which 80 per cent of the scores fall, and the twenty-fifth percentile is the point at or below which 25 per cent of the scores fall. 37 The distribution of scores falls along a normal curve in samples of average or large size. When each score is changed to a percentile equivalent, a small raw-score difference near the middle of the scale is equivalent to a larger percentile difference than one falling at either of the scale distribu¬ tion. For example, the difference between the fortieth and fiftieth percentiles represents a small difference in per¬ formance, whereas the difference between the ninetieth and ninety-ninth percentiles, although appearing small on the scale, may represent a great difference in performance or number of items correctly answered. In interpreting percentiles, the standard error of estimate, which is a function of the reliability of the test, must also be considered. It is thus best to interpret a person's percen¬ tile as a general range of percentiles rather than as a specific percentile. Use of Timed or Untimed Norms.— Norms were collected on the PRT so that it could be used as a timed or untimed test. As mentioned previously, maximum cultural fairness is achieved when the test is administered without a time limit, and this mode of administration is recommended whenever possible. Because there are situations in which it is difficult to utilize an untimed instrument, norms for a timed administration have been provided. 38 Use of Pictorial Reasoning Test in Industry The PRT is designed to be used as a general index of ability, measuring a person's potential to learn jobs inde¬ pendent of his background of culture. Seven subcultures took part in the normative testing of the PRT. The PRT was expressly designed to yield comparable measurement across subcultures. In the untimed condition there is little problem with comparability of subculture. Hence Table 3 provides only one general set of norms. If the test is to be timed, separate subcultural norms have been provided where there are differences. Table 4 will help to provide comparability of groups for inter¬ pretive purposes in the timed condition. In a manual it is unwise to give specific cutoff test scores that select persons who will succeed. Job demands vary greatly among companies, making it impossible to establish a set boundary of scores. It is suggested that those who use this test determine for themselves the approximate score levels for different types of jobs on the basis of cumulative experience with the test. Specific company norms should always be developed separately for each job group and may be a better tool than a verbal test for evaluating applicants for jobs requiring minimum reading skills. There are not 39 TABLE III. 1967 UNTIMED NORMS FOR THE PICTORIAL REASONING TEST Percentile Total Group 99 67 98 66 95 64 90 62 85 60 80 59 75 58 70 57 65 56 60 55 55 55 50 54 45 53 40 52 35 51 30 50 25 49 20 47 15 46 10 43 5 40 2 34 1 32 N 1250 X 52.8 S.D. 7.6 40 TABLE IV. 1967 TIMED NORMS FOR THE PICTORIAL REASONING TEST Percentile Total Group Control Group Urban White Other Groups Percentile 99 61 62 62 60 99 98 59 61 60 58 98 95 57 58 57 55 95 90 54 56 55 53 90 85 53 54 53 51 85 80 51 53 52 50 80 75 50 52 51 49 75 70 49 51 50 48 70 65 48 51 49 47 65 60 47 50 48 46 60 55 47 49 48 45 55 50 46 48 47 44 50 45 45 47 46 44 45 40 44 46 45 43 40 35 43 45 43 42 35 30 42 44 42 41 30 25 41 43 41 40 25 20 39 42 40 39 20 15 38 41 39 37 15 10 36 39 37 36 10 5 33 36 34 32 5 2 29 33 30 27 2 1 25 31 27 25 1 N 3598 1337 318 1943 N X 45.2 47.7 46.0 43.4 X S.D. 7.5 6.6 7.3 7.6 S.D. 41 specific norms for industry available in this preliminary manual; however, general industrial norms are being developed as guidelines and will appear in later manuals. Until local norms are established, the education norms may be helpful and should be interpreted in terms of a range of scores. It should be remembered that a test of this nature yields only an estimate of general ability. If a cutoff is established, applicants scoring within a standard error below this score should also be considered, utilizing other information ob¬ tained in the employment process. CHAPTER V. ANALYSIS AND INTERPRETATION OF DATA At the time of this study, only the NCR Personnel Battery Test was being administered to all applicants. It was ori¬ ginally selected as part of "Operation Upgrade" which was established in 1962 by the company as a result of a study by a management consulting firm. Available information indi¬ cates that the test was selected not necessarily for its application to specific jobs but for the purpose of eliminating a certain percentage of applicants from consideration for employment. While this objective is in keeping with the basic purpose of testing, the NCR seemed to be an unappropriate choice since it measures aptitude for checkers rather than the general ability of individuals. In addition to these factors, several administrative problems also arose in the use of the NCR Battery Test. Time.—The NCR Battery Test requires strictly timed test conditions in order to achieve accurate test results. Due to the nature of the test, the time allowed an applicant can radically affect the results. Generally, good test conditions are achieved when the test is given in the Personnel Office. However, seventy per cent of the company stores are forced to do their own testing due to distance from headquarters. The total time for administration of the NCR Battery Test is 43 thirty minutes. Tying up a store manager's or another adminis¬ trator's time is a significant hidden cost of test administra¬ tion. Due to this factor, and to achieve a degree of effi¬ ciency in testing applicants, certain times and places were designated for administration of the tests. This resulted in applicants losing interest in employment with the company and seeking work elsewhere. Availability of Test Results.—Generally, the scoring of the test was done by the Personnel Department. Frequently, the store manager did not know the results of the test when he made the decision to hire or reject an applicant. This resulted in some individuals being hired and, when their test results were known, having their employment terminated. Besides the obvious effect this has on public relations, the procedure promoted higher recruting and turnover cost for the company. The real value of testing in the selection process was lost in the evaluation of test results as one indicator of suitability for employment. After examination of various tests available to industry, the SRA Pictorial Reasoning Test (PRT) was selected for further study to determine its suitability to the company's needs. The principle reasons for selecting this test were: 1. Ease of administration. Instructions are written at fourth grade level of interpre¬ tation. 44 2. The test is untimed which releases the store manager from responsibility of timing the test. 3. The test gives immediate results to the store manager on persons responsible for hiring. 4. It meets many of the guidelines being advanced by the Equal Employment Oppor¬ tunity Commission since various sub¬ cultural groups that are likely to be applicants for employment with the company were part of the population used in establishing norms for the test. 5. Cost of this test compares favorably to the cost of the National Cash Register Test Battery. In order to determine the usefulness and effectiveness of the Pictorial Reasoning Test for Colonial Stores, sixty-five individuals were selected and tested. The first group consisted of thirty applicants and the second, thirty-five employees. Both groups were administered the National Cash Register Personnel Battery Test and the Pictorial Reasoning Test. Addi¬ tionally, the employees were evaluated with a specially developed Personnel Rating Sheet. See copy in the appendix. To further improve the accuracy in the selection process, a simplified form was devised, the Employee Selection Guide (see copy in appendix). The form provides space for name, date, location, and simple directions for completing the form. It is to be completed by the store manager and all information comprising the rating is confidential and is not to be revealed 45 to the applicant. It was the intention to make this form simple to complete but comprehensive enough to be an effective evaluative tool. Beginning at the right and moving to the left, the descriptions listed in the first two columns denote caution in hiring. Finally, the interviewer (store manager) is required to make an overall rating of the applicant at the bottom of the form. The last section determined if the applicant was acceptable for employment. If not, a detailed explanation of reason for rejection was required. The major purpose of the guide was to give store managers a tool to evaluate potential employees, encourage the review of all applicants when an opening was available and to provide an explanation for acceptance or rejection of an applicant. Description of Sample Groups.—Of the thirty applicants included in the sample for this study, twenty-four were male and six female. Of the twenty-four males, a total of seven represented minority group members. In the group of thirty- five employees, twenty-four were male and eleven were female. Of the total, eleven males and five females represented minority group members. A cut-off score of 400 on the National Cash Register Test Battery had been suggested to the company. The score of 52 or the fortieth percentile was selected as the cut-off on the 46 Pictorial Reasoning Test. Subsequent pages present the find¬ ings in the study. For purpose of discussion, the applicant group has been divided into two sub-groups: (1) those scoring less than 400 on the National Cash Register Battery Test; and (2) the applicants scoring 401 and above. Comparison of Scores on the National Cash Register Personnel Test and the Pictorial Reasoning Test.—In Table 5, composed of all applicants in the sample scoring less than 400, contains 46.2 per cent of the total applicant sample. In the method of employee selection previously utilized, these persons would probably have been eliminated. Of this number, 35.1 per cent scored 52 and above on the Pictorial Reasoning Test, which means that they were qualified for further consideration for employment provided they met other criteria. There were a total of sixteen applicants who scored more than 400 on the National Cash Register Battery Test. Of this number, eleven or 68.1 per cent scored 52 or above on the Pictorial Reasoning Test. See Table 3 for full summary. In reviewing the results of the performance on the two tests, an additional sixteen per cent qualified for considera¬ tion for employment by using the Pictorial Reasoning Test. These five individuals who had failed to achieve the cutting score of 400 on the National Cash Register Battery Test were found to possess the general ability to learn to perform within various job classifications in Colonial Store. 47 TABLE V. APPLICANTS SCORING LESS THAN 400 ON THE NATIONAL CASH REGISTER TEST AND RAW SCORE AND PERCENTILE RANKING ON THE PICTORIAL REASONING TEST Applicants National Cash Register Pictorial Reasoning Test Personnel Test Raw Score Percentile 1. CRC* 88 54 50 2. JCW* 184 40 5 3. LLC* 189 54 50 4. LHF 247 46 15 5. TLD* 257 45 14 6. DJH 260 40 5 7. JRJ 300 57 70 8. WCR* 315 40 5 9. HLA* 322 52 40 o 1—1 JWD 332 58 75 11. MG + 337 35 2 12. AVM + 358 46 15 13. CRA* 363 63 90 • i—1 RJH 397 43 10 * Negro + Female 48 TABLE VI. APPLICANTS SCORING 401 AND ABOVE ON THE NATIONAL CASH REGISTER TEST AND RAW SCORE AND PERCENTILE RANKING ON THE PICTORIAL REASONING TEST Applicants National Cash Register Pictorial Reasoning Test Personnel Test Raw Score Percentile 1. CC + 425 53 55 2. WPT 438 59 80 3. RDF 479 61 87 4. WOB 481 54 50 5. TDM 507 57 70 6. SKB + 535 48 23 7. WBM 544 62 90 8. SAE + 549 54 50 9. RCP 567 58 75 10. RCP 603 52 42 11. SGQ + 626 53 45 12. DEM 660 38 4 13. CLS 714 62 90 14. DLW 737 58 75 15. RJF 786 50 30 16. CRT 928 65 96 + Female 49 Employee Ratings on the National Cash Register Battery Test, Pictorial Reasoning Test, and Personnel Rating Sheet.— Of the thirty-five employees, nineteen or 54 per cent achieved the cutting score on the National Cash Register Battery Test. Of this number, nineteen, thirteen, or 70 per cent also achieved the cutting score on the Pictorial Reasoning Test. In the employee group, a total of seventeen or 48.2 per cent achieved the cutting score of 52 on the Pictorial Reasoning Test. See Table 4 for full summary. In addition to administering the two tests to the employee group, each person's supervisor was asked to evaluate his actual job performance. Thirty-four returns were received. Table 8 summarizes the ratings. TABLE VII. EMPLOYEES' RATINGS ON THE NATIONAL CASH REGISTER TEST, PICTORIAL REASONING TEST, AND PERSONNEL RATING SHEET Employee National Cash Register Pictorial Reasoning Personnel Personnel Test Raw Score- -Percentle. Rating 1. LM* 118 48 23 Good 2. JT* 132 39 5 Good 3. NM* 207 45 15 Satisfactory 4. HLE 209 55 60 Good 5. EC*+ 212 51 35 Good 6. AH*+ 274 52 40 Good 50 TABLE VII. - Continued. Employee National Cash Register Pictorial Reasoning Personnel Personnel Test Raw Score-Percentle. Rating 7. SEW+ 283 44 17 Poor 8. WLD 289 30 1 Good 9. CEB 307 51 35 Good 10. CEC 317 64 95 Satisfactory 11. WLM 336 57 70 Good 12. CW 341 50 30 Satisfactory 13. FWS 343 39 4 Good 14. MH*+ 362 51 35 Good 15. SW* 389 60 85 Satisfactory 16. AL* 392 58 75 Good 17. LRJ*+ 421 64 95 Good 18. ARE 435 40 5 Satisfactory 19. ALR 440 38 3 Satisfactory 20. LB* 455 55 60 Satisfactory 21. AAM 463 62 90 Satisfactory 22. LLG+ 466 64 95 Satisfactory 23. GFF* 475 49 25 Satisfactory 24. RLB* 477 65 95 Good 25. AAW+ 480 57 70 Satisfactory 26. MRC+ 483 48 23 ** 51 TABLE VII. - Continued. Employee National Cash Register Pictorial Reasoning Personnel Personnel Test Raw Score -Percentle. Rating 27. BWG+ 516 54 50 Good 28. NKP+ 521 58 75 Good 29. RBJ 526 48 23 Good 30. AT*+ 548 57 70 Superior 31. HA* 645 55 60 Good 32. HAS* 646 47 20 Good 33. JRB 656 53 45 Satisfactory 34. TMB 685 54 50 Satisfactory 35. TAN* 730 61 85 Good *Negro +Female **Personnel Rating Incomplete TABLE VIII. SUPERVISOR'S RATINGS OF EMPLOYEE'S JOB PERFORMANCE Category Number Per Cent of Total Superior 1 2.9 Good 19 54.2 Satisfactory 13 37.1 Poor 1 2.9 Incomplete 1 2.9 TOTAL 35 100 52 Of the thirty-three employees who were rated satisfactory and better, sixteen scored less than 400 on the National Cash Register Test Battery. This would tend to substantiate the long-standing belief that actual performance on the job is the real test of ability to do a job. While the value of a rating scale depends to a large extent on the understanding of the person completing it, there is no doubt that it has value in the spectrum of employee selection and retention. The scale developed by this writer was a beginning at development of a workable instrument to be utilized consistently within the company structure for employees. Much work needs to be done of the factors included and their relevance to actual job assignments and performance of these duties. CHAPTER VI SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Summary In recent years, increasing attention has been devoted to the utilization of tests in the employment selection process. Clearly, the usefulness of a test depends, among other factors, on the extent to which it measures what it purports to measure. A valid test is one that shows a significant relationship be¬ tween success on the test and success on the job. Testing is but one of the tools in the selection process. A test usually reveals information in only a single job area. Too often candidates are chosen primarily on the basis of their performance on a test without taking into account other criteria for employment. The purpose of this study was to describe the efforts of Colonial Stores, Incorporated, to re-evaluate its employment selection process in light of changing needs and conditions of employment. Specifically, it has examined the effectiveness of the National Cash Register Personnel Battery Test as a predictive measure of job performance. The use of the NCR Battery Test was begun in 1962 as a part of "Operation Upgrade." Available data indicates that the test was not necessarily selected because of its applica¬ bility to specific job assignments, but for the purpose of eliminating a certain percentage of applicants from further 54 consideration for employment. While this purpose is in keeping with goals of testing, the NCR Battery seemed an in¬ appropriate choice since it was constructed for use in screening prospective check-out cashiers. After examination of various tests available to industry, the Science Research Associates Pictorial Reasoning Test was selected for further study to determine its suitability to the company“s needs. Some of the reasons for its consideration were its ease of administration and scoring; it could be administered as an untimed test; it meets many of the guide¬ lines advanced by the Equal Employment Opportunity Commission as far as establishment of norms for various sub-cultures within the general population. In this study, sixty-five individuals were tested: thirty applicants and thirty-five employees. It was found that with the use of the Pictorial Reasoning Test, 35.1 per cent more applicants were available for employment consideration based on test performance alone than would have been the case with the NCR Personnel Battery. In the employee group, of the nineteen who achieved the cutting score of 400 on the Personnel Battery, 70 per cent also achieved the cutting score of 52 on the Pictorial Reasoning Test. Additionally, the employees were rated by their super¬ visors by use of a rating scale devised by this writer. Of the thirty-three employees who were rated satisfactory and 55 above, sixteen scored less than 400 on the NCR Battery. This finding would tend to substantiate the long-standing belief that actual performance on the job is the real test of ability to satisfactorily execute an assignment. Conclusions Individuals are of primary importance in a free society, where each man has a right to a job for which he is qualified -- as long as he fulfills its demands. Some men in trades or a profession may work independently, but most will find employment by joining some organization. In a free society, organizations are formed when people voluntarily join together to reach common goals. They earn this privilege by producing goods or services of value to at least a segment of the popu¬ lation. A legitimate organization deserves, therefore, to 1 prosper and grow according to its contribution to society. As it grows, it adds new members for certain functions and responsibilities. The interests of society, as well as of the organization, are served best when the new members are fully qualified for their jobs. The interests of society are compromised when a fully qualified person is denied a place ‘ 1 'Eells, R. , and Walton, C. , Conceptual Foundations of Business (Homewood, Illinois: Irwin Press, 1961), p. 63. 56 given to one less qualified. If many serious errors of selection are made, moreover, the organization may fail — with resulting human and economic waste. The decision to apply for work is made by the individual, but the decision to hire or not is made by the organization. The point of view of this writer is that it is both wasteful and immoral to deny qualified persons desirable and available employment for invalid reasons. Tests and test technology used in employment procedures can markedly reduce the fre¬ quency and degree of wasteful selection errors. The employing organization has a responsibility to itself and to the society that supports it to be sure that it uses these tools competent¬ ly and wisely. Recommendations Tests cannot be relied upon as the sole criteria for selecting or rejecting an applicant. Companies that have come to grips with the problem of inadvertent discrimination through the improper use of testing procedures have developed a number of ideas on how to avoid the pitfalls of testing. These ideas suggest areas in which those persons engaged in the employment selection process can specifiy their work. On the basis of the interviews with personnel managers, job applicants, and others, knowledgeable and interested in the employment selective process, the following recommendations 57 appear to be relevant: Determine and state in specific terms the exact nature of the abilities necessary for the job.—Clearly indicate what equipment or tools the employee will be required to use; the level of arithmetic he must know; the type or writing he will be required to do; the kind of instructions he will have to understand; the types of problems he will face. Written job requirements will facilitate choosing the correct selection procedures and also will tend to limit subjective judgments and bias in interviewing. Make sure that the state job requirements are essential to performance on the job and do not call for vague and arbitrary standards such as "must be intelligent" or "must be high school graduate" or "must have potential to advance to higher level." Social acceptability should not be confused with capability to do the job. Choose tests on the basis of specific job-related criteria.—Make sure the tests are valid for the individual job and setting and relate to the applicant's ability to per¬ form the desired function. Make active effort to seek out and employ "qualified" minority group applicants.—An intensified recruting effort is justified because minority group members, as a result of discouraging experiences in the past, are often hesitant to apply for jobs they actually are capable of filling successfully. 58 The screening and interviewing of minority applicants should be conducted by personnel thoroughly committed to the policy of equal employment opportunity as well as knowledge in intergroup relations.—Employment personnel should be aware that minority applicants are often especially sensitive and apprehensive in the employment situation. Many feel that the tests are designed to exclude them; they may have had little experience with written tests, and they may have pro¬ blems with the English language. They may also dress and look differently, and may appear less knowledgeable to the unini¬ tiated. Yet they may be fully productive workers in many jobs. The interviewer should be able to communicate his own and his company<1s sincere desire to treat all applicants equally. He should be able to put the applicant at ease so that he will give a true picture of himself and his abilities. Do not set unrealistically high test-score standards that will lead to the hiring of over-qualified employees.—This can result in low morale on t he job on the part of the over- qualified person who may not be sufficiently stimulated by the work he is doing and therefore can lead to higher turnover rates and increased recruting costs. Make sure tests are not screening out those applicants (particularly minority group members) who are capable of performing effectively on the job, but whose cultural or 59 economic backgrounds handicap them in taking the tests. Use testing as only one indicator among others in the hiring decision.—There should be a clear awareness that where the applicant has not shared in the predominant middle-class verbal culture, the test score may significantly underestimate his potential. Personal characteristics such as achievement, motivation, and dependability may be even more important than test scores in indicating successful job performances. Look at the whole individual, not just at one specific characteris¬ tic, in making a decision. Tests should be developed by reputable professional psychologists who are competent in conducting testing programs in an industrial setting. Retesting should be offered to applicants who have availed themselves of the opportunities for more training or experience. Since the disadvantage that minority group applicants often face in the testing situation, they should be given repeated chances, within reason, to demonstrate their real capabilities. For many jobs, the difference between an unqualified applicant and a qualified one may be a modest amount of training.—Funds are available to employers who will assist the unemployed to become qualified through training. Consideration should also be given to the possibility that just a little more learning time on the job might result in a considerably better 60 performance on the part of applicants whose backgrounds lacked relevant learning opportunities. APPENDICES 61 APPENDIX A PROCEDURE FOR CENTRALIZED HIRING OF STORE EMPLOYEES Responsibility Action Employee Relations Department 1. ALL APPLICANTS must complete an application and health questionnaire and be inter¬ viewed. The Employee Selec¬ tion Guide should be completed by the interviewer. 2. If UNACCEPTABLE for employ¬ ment, give detailed reasons on the Employee Selection Guide and file with the other employment forms in the "Inactive Employment File" for a minimum of six months. 3. APPLICANTS must be tested before hiring. 4. If ACCEPTABLE, file the application and other employ¬ ment forms by data within job classification in the "Active Employment File" for further consideration. After ninety days remove and file in the "Inactive Employment File. " 5. A FILE should be maintained on all existing vacancies, date job order was placed, store number, job classifica¬ tion, and the date of re¬ ferral . 6. When a VACANCY arises, the applicant best qualified is selected from the "Active Employment File" and re¬ ferred to the store. (Use C.S. Form-6249). APPENDIX A Continued. Responsibility Action Store Manager Employee Relations Department 7. When an applicant is REFERRED but not hired, detailed rea¬ sons should be given on the Referral Card and forward immediately to the Employee Relations Department. 8. When an applicant is REFERRED and hired, the Referral Card should be completed with tax forms and a Form 15, then forward to the Payroll Depart¬ ment with the current week's payroll. 9. ALL APPLICANTS applying for jobs at the store must be referred to the Headquarters Office for processing and testing. 10. Complete WRITTEN reference checks. 11. Within the PROBATIONARY PERIOD, review all employment forms and notify the store manager of any irregularities. APPENDIX B HIRING Responsibility Store Manager Employee Relations Manager PROCEDURE AT STORE LEVEL Action 1. ALL APPLICANTS must complete an application and health ques¬ tionnaire and be interviewed by the store manager. The Employee Selection Guide should be com¬ pleted by the interviewer. 2. If ACCEPTABLE but not hired, file the application and other forms by date within job classification in the "Active Employment File" for further consideration. After ninety days remove and forward to the Employee Rela¬ tions Department with the nota¬ tion, "Inactive Employment File." 3. APPLICANTS must be tested before hiring. 4. If an applicant is HIRED, forward the application, employment and tex forms with a completed Form 15 on top to the Payroll Depart¬ ment with the current week's payroll. 5. If an applicant is not ACCEPTABLE give detailed reasons on the Employee Selection Guide and for¬ ward along with all employment forms to the Employee Relations Department immediately. 6. Refer ALL APPLICANTS who apply at Headquarters to an appropriate store for interviewing and processing. 7. Approve in advance ALL advertis¬ ing for vacancies. APPENDIX B Continued. Responsibility Employee Relations Manager Action 8. Establish an "INACTIVE Employ¬ ment File" for applications of unacceptable applicants and applications over ninety days old. Applications should be held in the "Inactive Employ¬ ment File" for a minimum of six months. 9. Complete WRITTEN reference checks. 10. Within the PROBATIONARY PERIOD, review all employment forms and notify the store manager of any irregularities. APFEKDEC C . Form 6243 EMPLOYEE SELECTION GUIDE . 10/67 NAME OF APPLICANT LOCATION OR STORE H DATE PLACE AN "X" IN THE BLOCK THAT BEST DESCRIBES THE APPLICANT. WHEN COMPLETED, ATTACH TO THE EMPLOYMENT APPLICATION. Formal Did Not Graduate Now Attending High High School Completed One or Education From High School School or Trade Graduate More Years School Graduate College □ n □ □ Health (Visual Observation Unacceptable Poor Acceptable Good & Health estionnaire) □ □ □ □ Related None One Year or Less One to Two Over Two Years Experience Years □ □ □ □ PRT Test Below 52 52 to 55 56 to 60 Over 60 Results □ □ □ □ Age * Over 41 Under 16 31 to 40 16 to 25 26 to 30 - □ □ □ □ , Phone Check Last Local Unacceptable Did Not Check Acceptable Good Employer .□ □ □ □ Only . Appearance Undesirable Marginal Acceptable Good Neatness & Orderl ines s □ □ □ □ VERALL EVALUATION (DETERMINE FROM "X” MARKS ABOVE CCEPTABLE FOR EMPLOYMENT Yes No (If no. give detailed reasons) COMPLETED BY • If acceptable but not hired, file with application and other employment forms in the Active Employment File. • If hired, forward with the application and other employment forms to the Payroll Department with the current week's payroll. '• If not acceptable, give detailed reasons and forward with all employment forms to the Employee Relations Department immediately. ♦ Age not to be considered in Ohio and Maryland. APPENDIX C Form 6243 EMPLOYEE SELECTION GUIDE 10/67 NAME OF APPLICANT LOCATION OR STORE # DATE PLACE AN ”X” IN THE BLOCK THAT BEST DESCRIBES THE APPLICANT. WHEN COMPLETED, ATTACH TO THE EMPLOYMENT APPLICATION. Formal Did Not Graduate Now Attending High High School Completed One or Education From High School School or Trade Graduate More Years School Graduate College □ □ □ □ Health Visual bservation Unacceptable Poor Acceptable Good . Health stionnaire) □ □ □ □ Related None One Year or Less One to Two Over Two Years Experience Years □ □ □ □ PRT Test Below 52 52 to 55 56 to 60 Over 60 Results □ □ □ □ Age * Over 41 Under 16 31 to 40 16 to 25 26 to 30 □ □ □ □ Phone Check ^' Last Local Unacceptable Did Not Check Acceptable Good Employer .□ □ □ □ Only Appearance Neatness & Undesirable Marginal Acceptable Good Orderliness □ □ □ □ HALL EVALUATION (DETERMINE FROM "X” MARKS ABOVE JEPTABLE FOR EMPLOYMENT Yes No (If no. give detailed reasons) COMPLETED BY ■ - If acceptable but not hired, file with application and other employment forms in the Active Employment File. If fakired, forward with the application and other employment forms to the Payroll Department with the current week's payroll. If act acceptable, give detailed reasons and forward with all employment forms to the Employee Relations Department immediately. * Age not to be considered in Ohio and Maryland. personnel test For Prospective Check-Out Cashiers NCR Developed For The National Cash Register Company Dayton, Ohio By Ward J. Jenssen, Inc. Los Angeles, California © Copyright 1961 by The National Cash Register Company Dayton, Ohio, U.S.A. ♦TRADE-MARK —REG. U. S. PAT. OFF. PERSONNEL TEST FOR CHECK-OUT CASHIERS PART I Irthis test you will be asked to check two sets of cash register tapes to see if they are the same or different. Place an "S" in the answer column every time two lines on the sample tapes are exactly the same; a "D" if they are different in any way. When checking the cash register tapes do not worry about the arithmetic. Do not total any of the tapes to see if the addition is correct. You are to check only for errors in corresponding lines. Example: Please do the last four items. Mac's s Mac's 12 Main St. s 12 Main St. .29 P D .29 G .16 G S .16 G 1.25 M S 1.25 M 6.09 M D 6.90 M 7.79 TOTAL D 7.97 TOTAL Joe's Stores D Joe's Store No. 5 D No. 9 10.07 M D 10.97 M .25 P D .25 G .17 G 1.98 P .54 G 13.01 TOTAL .17 P 1.87 P .54 G 13.00 TOTAL The correct answers should be: D D S D YOU WILL HAVE EIGHT MINUTES FOR THIS TEST. DO NOT TURN THE PAGE UNTIL YOU ARE TOLD TO DO SO. Perce's Perce’s No. 626 No. 628 Nov 11 085 Nov 11 058 .10 P .10 P .19 P .19 G 4.52 M 4.25 M .23 G .23 G .87 G .87 G .63 G .62 G 1.18 P 1.17 P 2.95 G 2.95 M .96 M .96 G .66 P .66 P .72 P .72 P .27 G .27 G .28 G .26 G 11.01 M 11.01 M 2.18 M 2.18 M .69 P .60 P .99 P .99 G 1.08 G 1.08 G .50 P .05 P .08 P .08 P .81 M .81 M .16 G .16 G .37 M .37 M .38 M .33 M .33 G .38 G .48 M .49 M 4.12 M 4.12 M .41 G .14 P .41 P .41 P Total 37.56 Total 36.56 Thorp Nolan Throp Nolan Supermarket Supermarket 8.81 M 8.81 M .45 P .49 P .68 P .66 P .92 G .92 G 1.81 M 1.81 M 2.54 G 2.53 G 3.00 P 3.00 G .16 P .16 P .71 M .17 M .31 G .31 G Total 19.39 Total 19.39 W. Brown Co. W. Browne Store No. 196 Store No. 196 4.56 M 4.56 M .38 G .38 G .69 P .69 P 8.91 G 8.19 G 5.90 G 5.09 G 1.53 M 1.53 P .16 P .16 P .73 P .73 P .39 M .39 G 1 .42 P 42 P .89 M .87 M .19 G .19 G .61 G .16 G Total 25.36 Total 25.36 Sid Johnson Syd Johnson No. 147 No. 147 .25 G .25 G .61 P .60 P .39 M .38 M .73 M .73 M 2.22 G 2.22 M 1.85 G 1.85 G Total 6.06 — Total 6.06 Spencer Foods Spencer Food 122 S. Marten 124 S. Martin 9.20 M 9.20 M .26 M .26 P .84 G .84 G 1.01 G 1.10 6 .56 M .56 M .28 M .29 M 6.74 G 6.71 G 4.82 M 4.82 M .19 P .19 P .12 P .12 P 2.01 G 2.10 G .43 P .43 G .94 G .94 G Total 28.56 Total 27.56 PART I Page I GO ON TO THE NEXT PAGE Clagcr & Son Clager & Son Good Food Good Food Store #72 Store *72 1 134 Geneva 1314 Geneva 6.71 P 6.7.1 P 5.01 M 5.01 M 2.11 S 2.11 M .79 G .77 G 3.05 M 3.05 M .23 P .23 G .23 G .23 P .39 G .39 P .05 G .50 P .41 G .41 G .75 P .75 G .78 M .87 M 4.50 G 4.50 G 1.11 G Ml G 4.12 M 4.12 M 6.64 M 6.46 M .17 G .17 P 2.10 M 2.12 M .17 P .17 G .15 G .15 P .71 G .71 G .20 P .20 P .95 M .95 G .89 G .89 G 3.57 M 3.57 M .76 G .76 G 4.98 G 4.89 G Total 19.46 Total 19.46 .65 P .65 G .84 G .84 G Total 33.56 Total 33.56 Best Butcher Best Butcher 15 S. State St. 155 State St. Jerome's Jerome' 10.50 M 10.50 M 51 1 First St. 591 First St. 6.52 M 6.52 M 1.50 P 1.50 M .51 P .51 P 2.11 P 2.11 P .91 G .19 G .87 G .87 P .25 M .25 P .59 G .59 G .95 P .95 M .21 G .21 G .15 M .15 M 3.75 M 3.75 M 1.11 G Ul G .19 G .18 G .87 P .87 P .18 G .19 G .Total 4.75 Total 4.57 Total 26.42 Total 26.42 L. Franklin Co. L. Franklin Co. Simons Simmons Store #3 76 Store #376 576 Main 576 Main 5.07 M 5.07 G 1.69 M 1.69 P .21 G .21 G 3.75 M 3.57 M .17 G .17 G .89 P .89 P .87 P .78 P .26 G .26 G 4.02 G 4.02 P .57 G .56 G .54 P .54 G 2.39 G 2.39 G .32 P .32 P 2.63 M 2.63 M .18 G .19 G .25 P .25 M .24 G .24 G .56 P .56 M .24 P .24 G 5.41 P 5.14 P 3.06 M 3.06 M 9.06 M 9.06 M 4.58 G 4.87 G 1.28 G 1.38 G 6.11 M 6.11 M .15 G .15 G Total 25.61 Total 27.61 Total 29.98 Total 29.89 RT I Page 2 GO ON TO THE NEXT PAGE Washburn Store Washburn Store No. 251 No. 251 5.55 M 5.55 M .85 P .85 P .75 G .74 G 1.51 G 1.15 G 2.14 M 2.14 G .16 P .16 P .25 P .25 P 1.19 M 1.19 G .78 G .78 M .95 M .95 M 3.91 P 3.91 P 4.05 M 4.05 G 2.20 G 2.21 G .38 P .38 P .76 G .76 P 6.56 M 6.56 G .62 M .62 M .81 D .81 D 4.30 G 4.03 G 1.23 P 1.32 P .99 G .99 P .45 P .46 P 7.89 D 7.99 D 3.69 M 3.69 M 2.01 M 2.01 M 5.85 G 5.58 G .27 P .27 M Total 31.63 — Total 31.63 Getz Bros. Getez Bros. & Company & Company *85734 — *85734 8.53 M 8.52 M .80 P .80 P .21 P .21 P .94 G .94 G 3.21 G 3.21 G 10.01 M 10.10 M 4.01 M 4.01 P .77 G .78 G .64 G .61 G .77 P .77 G 2.90 M 2.91 M 5.32 P 5.32 P 9.31 M 9.31 M Total 47.72 Total 47.42 Joseph's Joseph's No. 725 No. 725 Jan. 6 795 Jan. 6 795 .15 P .15 G 2.58 M 2.58 M .85 G .85 G 6.76 M 6.67 M 3.43 M 3.43 M .15 P .15 G .25 P .25 G .87 M .87 M .98 G .98 G 1.89 G 1.98 G .18 G .18 G .11 P .11 P .25 P .25 P .53 M .55 M .55 P .53 P .53 G .58 P .58 G .53 G .65 G .65 G 1.49 M 1.49 M .91 P .91 M Ml M 1.11 M .75 G .75 G .22 G .22 G .19 P .19 P .42 D .42 P .73 P .73 D 2.93 M 2.93 M .59 G .59 G .68 M .68 M 1.86 P 1.86 P 1.19 P 1.16 P Total 25.96 Total 25.69 Temple Inc. Temple Co. 592 Bordon 592 Bordin .26 P .27 P .17 G • .17 G .82 G .82 G 2.81 M 2.28 M .51 M .50 M 2.31 P 2.31 P 4.13 G 4.31 G .90 G .90 G Ml M 1.11 M .37 P .37 G .64 G .64 G .99 P .99 P 5.59 M 5.95 M Total 20.61 Total 21.61 ,, ^ PART I Page 3 STOP, END OF PARI PERSONNEL TEST FOR CHECK-OUT CASHIERS PART 2 PLEASE READ THE INSTRUCTIONS ON THIS PAGE CAREFULLY. FOLLOW THESE INSTRUCTIONS EXACTLY WHEN DOING THE TEST PROBLEMS. DO NOT TURN THIS PAGE UNTIL YOU ARE INSTRUCTED TO DO SO. This is a test of problem solving ability. There are various types of questions in this part of the test. Here is an example of a type of question you will find. What number should be next in this series 2 4 6 8 ? 10 The correct answer is "10". So this number is written on the line at the right. Answer the next question yourself. The figure shown as "B" is correct so you should have written "B" on the line at the right of the page. Here is another example: Which number is the largest 3.9 1.995 .999 13.91 The correct answer is 13.91 so you should have put that number on the line at the right. In this test, when the answer to a question is a number or letter, put the letter or number on the line at the right. You may mark the paper in any way, but be sure your answers are shown in the answer column. Speed and accuracy are both important in this test. If you cannot answer one question, skip it and go on to the next. You may not finish all of the problems but work as quickly as possible. Ask questions about anything that you do not understand before the examiner starts timing you; no questions can be answered after the test begins. You will have ten minutes for this test. DO NOT TURN THE PAGE UNTIL YOU ARE TOLD TO DO SO 1. What number should come next in this series? 5 6 8 9 10 12 ? 2. Which number is SMALLEST? 4.2 3.1 3.01 2.3 2.4 3. / ./ is to / / as | ,| is to: A. |. | B. /’ / C./\ D. |' | 4. Which figure should come next in this series? 5. Which number is SMALLEST? 1. .89 .801 .99 1.9 6. F isto r as F isto: A. J_ B. ~| C.A D. 7. What number best follows this series? 42 39 36 33 ? 8. Which number is LARGEST? 349.892 100 892.3 7659 765.9 9. | | is to | • | as is to: A. B. —-— C. | D. —| Using the numbered symbols below, do the indicated arithmetic problems and place the correct answer in the column at the right. LLQAHlk^r7!^ ZZ7 'o- O + A + L - □ + A = ii- 0 - L + A + ^ - O = 12. * A + n + 13. □ + A X O - O 14. a + L + L - □ * o = 15. Which figure should come next in this series? 1 A. I* B. 1 C. AN D. | PART 2 Page I GO ON TO THE NEXT PAGE 16. Which number is the LARGEST? 90. 52 90. 90 82. 9 20. 0 5. 3 17. o^l is to — as DQ is to: A. oQ B. CO C. ^ D. 18. What number should come next in this series? 12 1 13 2 14 3 15 ? ;9. If the following numbers were arranged in a series, from the smallest to the largest, which number would be in the middle? 6493 152 5286 70. 53 1.93 20. is to □ as C is to: A. B. o c0 D. 21. Which figure should come next in this series? A. B. C. D. 22. Which number is the SECOND LARGEST? .92 .831 .05 .10 is to: A. 0/ /o n 23. Z is to / as - /O /O 24. What number should come next in this series? B. .794 1 □ C. m D. 3 2 1 6 5 4 9 ? 25. Which number is the SECOND LARGEST? 1. 2 .0003 .0021 . 1591 1.054 26. o is to OO as | | is to: A. □□ B. □ n r Q *0 Using the numbered symbols below, do the indicated arithmeti c probl ems. Place the correct answer in the col umn at the ri ght. | 1 © /3\ 0 IK $ 0 Zi/ 27. ZG - n + G + o - A = 28. L_ - 0 + A - O - □ = 29. - o ^ t\ - I_ IT 30. n + o - n - GG = 31. A + □ + □ - A = ANSWER COLUMN PART 2 Page 2 GO ON TO THE NEXT PAGE 32. Which figure should come next in this series? ANSWER COLUMN A. B. C. D. 33. If the following numbers were arranged in a series from the smallest to the largest, which number would be in the mi ddle ? 1.953 .7654 8.91 52.43 120.5 54. is to Cj as is to: A. ^ B. ^ C-(T1 d -D 35. What number should come next in this series? 1 3 6 10 15 21 36. Which number is the SECOND LARGEST? 18000 18.999 37. is to ^ as 8543 17952 6685 is to: A. |\| B. 2 CW D- 38. Which figure should come next in this series? A. B. V c. D. 39. Which number is the second SMALLEST? 6.3 1.2 1.5 .53 4.3 40. is to as | | is to: A. B. C. D. 41. Which number is SMALLEST? 4.6 1.76 3.02 1.16 2.1 42. What figure should come next in this series? A. ^7 I c D •a 43. What number should come next in this series? 44. 8 7 12 11 10 15 ? is to A- Q B''^ D / is to/^/as i: GO ON TO THE NEXT PAGE PART 2 Page 3 ANSWER COLUMN 45. Which number is SMALLEST? .45 .405 .50 2.0 46. What figure should come next in this series? □ O *'C© B. D. LJ 47. What number should come next in this series? 1 3 3 5 5 5 7 ? is to ^ osC is to * A. B. _ c.o D. _ 48. 49. Which number is LARGEST? 753.921 100 841.1 9256 925.6 50. Which figure should come next in this series? A\ A. < > B. N/ D.. 51. What number best follows this series? 28 24 20 16 ? 52. is to (T) as is to A. 53. Which number is the LARGEST? 16.35 16.80 3.00 9.6 16 54. Which figure should come next in this series? A.-\- + x^ -x^ - + - - +x c -+-^ x- B.-+X4- ^•-7 1 X 55. What number should come next in this series? 15 2 16 3 17 4 18 ? PART 2 Page 4 STOP, END OF PART 2 PERSONNEL TEST FOR CHECK-OUT CASHIERS PART 3 i ne columns on the following pages represent keys on a cash register. Above each set of cash register • eys is shown a sales figure for that register. Using a pencil, mark the sale amount for that set of - eys. Start from the left and work to the right; work as quickly as you can. If you make a mistake md want to correct your answer, do so by circling the cash register key that you marked incorrectly .ind mark the correct key. '.Kample I: $.92 Example 2: $.88 In example I the sale figure was $.92 so the 90 key and the 2 key were marked out. You do Example 2. Any questions about the test should be asked before the examiner starts timing you. You will have two minutes for this test. DO NOT START UNTIL YOU ARE TOLD TO DO SO PART 3 Page I GO ON TO THE NEXT PAGE -> 4 _ M) 83 (!) $. 14 0 © $.77 0 © $. 79 0 © $. 68 0 © To) V G 0 © 0 © 0 © 0 © '?o^ G 0© 0© 0© 0 © To) G 0© 0 © 0© 0 © To) 0 0© 0 © 0© 0 © 40) V 0 0 © 0© 0 © 0 © To'i 0 0© 0© 0© 0© ! 0 0© 0© 0© 0© ; io': <0 0© 0© 0© 0 © $.96 (M) ® ® ® (™)© ® ® ® ® ® ® ® © @© © ® $.57 © © © © ©0 0© © © © © @® @© ® © i,. 49 i?0) (?) $. 38 0 © $.85 0 © 0 ® 0® 0 © © ® 0 © 0 © 0 (?) 0 © 0 © 0 © 0© 0 © © © 0 © 0 © 0 © 0© ® © 0 © 0© 0 © 0 © 0 © 0 © $.12 $.34 © © © © © © 0 © 0© 0© 0© 0© © © 0 © © © 0 © 0® 0® @© 0 © 0 © 0 © $.44 $.78 $.19 0 © 0 © 0 © 0 © 0 © 0 © 0© 0© 0© 0 © 0 © 0 © © © © © © © © © © © © © 0© 0© 0© 0© 0© 0© 0® 0® 0© $. 89 $. 48 i © (?) ® © ':0© 0© ;®© ©© (0© 0 © •0© 0© ;©© ©© (0© 0© :©© @© c0© 0© t PART 3 $. 24 $. 67 © © 0 © © © 0 © 0 © 0 © 0 © 0 © 0 © 0 © © © 0 © 0 © 0 © 0 © 0 © 0 © 0 © $. 86 $. 16 0 © 0 © 0® 0 © 0 © 0 © ® © © © © © © © © © © © 0 © 0 © @© @© 0 © 0 © $. 29 $. 87 0 © 0 © 0 © 0 © 0 © 0 © 0 © 0 © 0 © 0 © © © © © 0 © 0 © 0© @© 0 © 0 ® Page 2 GO ON TO THE NEXT PAGE PART 3 Page 3 GO ON TO THE NEXT PAGE Page 4 GO ON TO THE NEXT PAGE ® © ® © © © ©© @© ©© $.95 © ® © © ©© © © © © © © © © ® © © © © © © © © © © © © © ® © ® ® ® © © © PART 3 $.18 ® © $.19 ® © $.24 ® © $.57 ® © $.61 ® © $.31 ® © (*)§ © © ® © ® © © © © © © © © © @© ©© ©© ©© © © ® © ® © ® © ® © © © © © m © © ® © © © ® © © © © © ®ri © © © © ® © © © © © ® © ® © © © ® ® ® © ® © © © @(i @© ® © @© @© @© ® © ®(? © © © © © © © © © © © © ®(i $.75 $.48 $.85 $.13 $.74 $.33 $.52 ® ® ® © ® © © ® ® © ® © ® © © © ® © © © © © © © ©© © © © © @© @© @© © © © © © © © © © © © © © © © © © © © © ® © ® © @(! ® © ® © ® © © © ® © © © © © © © © © ® © ® © ® © @© @© © © @© © © © © © © © © © © © © © © © © $.66 $.14 $.98 $.83 $.51 $.86 $.2C ® © ® © © ® © © © © © © ®(l ® © © © © © © © © © ® © ©© © © © © ©© © © @© © © ® © © © ® © © © © © © © © © © © © © © © © © © © © © ® © ® © @© © © ® © © © © © © ® © © ® © © © ® © @© ® © @© © © © © © © © © © © © © © © ©a Page 5 STOP, END 0PJ5 Copies of this form, Number B-52-A, may be purchased from fhe local office of The National Cash Register Company. -H CO > Q_ 3 3 CO Q_ cr -< O X m O a O > GO X m 73 to -o m 73 to o Z z m to X m o z > r~ O > to X 73 m to 73 O O X TJ > “n o ^3 > o o TO m to to a > —i m Z > X m O o 3 T> a 3 ~< Method for Obtaining Raw Score RAW SCORE METHOD 1 Using Percentiles METHOD II Using Weighted Total Score Part 1 - Total Answered minus (Wrongs X 2) = Raw Score From Table 1 Raw Score X 1 = Part 2:=Rights only (Problems) From Table II Raw Score X 10 = Part 3 = Rights only (Register Keys) From Table III Raw Score X 2 = Personal Opinion Ballot (Questions) Scored by use of special stencil From Table IV Raw Score X 2 = Total of Weighted Scores (Compare with cutting scores that company has established) SRA Pictorial Reasoning Test Robert N. McMurry, Ph.D. Phyllis D. Arnold, M.S. Science Research Associates, Inc. A Subsidiary of IBM Copyright © 1966, Science Research Associates, Inc. All rights reserved. Printed in U.S.A. Self-Scoring Edition Name Chicago General Directions This is a test to see how carefully and quickly you can observe and reason. In the rows of pictures that follow, you will find that four of the pictures in each row are related in some way. Think of the way in which the four are alike. Notice that the one that is marked is different from the others in the row. In each of the four rows below, look at the pictures that are not marked and try to find the way that they are alike. 8^. SP is E ^ S 2 — m 1 — E 4 1! —^ E ram dS* ^P| few® tl v m v [g Sample Problems In the first row, four of the pictures are of musical instruments. The record player is different be cause it does not produce music directly the way a musical instrument does. An X has beer marked in box B beneath the record player. Notice that it would have been wrong to mark picture C because it is the only one that you beat, or to mark picture E because it is the only one with strings. All four pictures except the phono¬ graph are alike because they are musical in¬ struments, and ONLY the phonograph repeats the sounds made by other instruments. In the second row the D square with the figure 4 in the center has been marked. Look carefully at the other four squares. In each, the number is placed in a corner of the square. The fourth square is different because it is the only one with a number centered within it. In the third row. box D beneath the fourth man has been marked because each of the other four men is using his hands and making a motion that can be seen by others. Picture D shows the only man using his voice. Sample row 4 has five six-pointed stars Two of the points in each figure have been darkened. In four of the figures the two black points are directly opposite each other. In figure C the black points are not directly opposite, sc C is marked as the right answer. SRA Pictorial Reasoning Test Practice Problems Look at the six rows of pictures below. You are to work these problems to help you understand how to do the test. First find the four pictures that are alike in some way, and then mark the one that is not like the others. Place an X in the lettered box of the picture you choose. © o 0 0 0 olO' El © 5 0 m * 0 o is ^^El iP^ls * * •_MS / x r E X d El You should have marked box E in the first row because the two small figures are separated by the large figures in the other pictures, but here the small 0 and 1 are together. Box E in row 2 should be marked because the rest of the boats are moved by something other than motor power. Box A in row 3 should be marked because the other four have two black sections and this figure only has one. Box C should be marked in row 4 because all the other pictures contain one child and one adult. The pair in picture C are both children. In row 5, E should be marked because the other objects are used in indoor games, and the bat and ball are used outside. Box E in row 6 should be marked because the other items can be made smaller or larger in some way, while this one cannot. When you are told to begin, you will mark one picture in each row on the next four pages. Try to find the way in which four of the pictures in each row are alike, and choose the one that is different. Carefully mark each choice with an X by marking heavily. Do NOT mark anything other than your answers. If you want to change an answer, draw a circle around the box like this: (x)) . Then mark your new answer. Be sure you understand how to work these problems. If you have any questions, ask the person giving the test. Work quickly, but try not to make mistakes. Select only ONE picture in each row. STOP HERE-DO NOT TURN THE PAGE UNTIL THE EXAMINER TELLS YOU TO DO SO. Tu rn pa ge an d go o n Name N um be r c o rr e c t 1— 40 Tu rn pa ge an d go o n Reorder No. 7-576 APPENDIX F PERSONNEL RATING SHEET NAME _ , . ■ PRESENT POSITION - . DATE STORE tt DATE OF EMP* NAME OF RATER _ Prepare this rating carefully and accurately. Judge each characteristic or trait separately or independently from any of the others. Place an "X" in the square that best describes the employee being rated. Keep in mind that (1) represents Unsatisfactory (2) Poor (3) Satisfactory (4) Good and (5) Superior* v 1. Quantity of Work 2. Quality of Work 3. Knowledge of the Job 4* Ability to Learn 5. Willingness to Learn 6. Follows through on Instructions 7* Initiative 8* Interest 9* Gets along with Employees 10. Gets along with Customers 11. Gets along with Supervisors 12. Dependable 13. Exercises Good Judgment 14* Likes this type of Work 5 4 3 2 1 5 4 3 2 1 OVERALL RATING SIGNED BY APPENDIX G PERSONNEL RATING SHEET TO BE UTILIZED IN VALIDATION STUDY Instructions; Check one letter which best represents the amount, or quantity of the attibute listed. The letter a^ represents the greatest or most of the quality or charac¬ teristic, and the letter d represents the least or smallest amount. Remember to rate as truly as possible and don't fall into the trap of bad ratings. 1. Demonstrated skill is efficiency in completing tasks. a. Completes tasks with superior skill. b. Completes all but the most difficult tasks. c. Demonstrates poor or insufficient skills. 2. Initiative is talent for having new ideas, finding new and and better methods for job performance. a. Displays outstanding initiative in planning and completing work assignment. b. Above average interest, does more than is necessary. c. Shows average interest in work. d. Rarely shows interest, must be prodded. Accuracy is the correctness of work assignments performed a. Makes very few mistakes; corrects ! well as the errors of others. his own ! b. Makes few errors; corrects his own other's errors. as well c. Makes an average number of errors; corrects his own errors. usually d. Makes many errors; rarely corrects other's errors. his own 4. Knowledge of the job is information that should be known for satisfactory performance. a. Learns assignments quickly and well. b. Learns assignments fairly quickly. APPENDIX G - Continued. c. Has average goals, usually puts forth efforts to achieve them. d. Poorly defined goals, puts forth practically no effort. Stability is the ability to withstand pressure and remain calm in crisis situations. a. Thrives under pressure, really enjoys problem¬ solving . b. Tolerates most pressure, likes crisis situations more than average person. c. Has average tolerance for crises; usually remains calm. d. Ineffective under pressure; nervous. Alertness is the ability to grasp instructions, to meet changing conditions, and to solve novel problem situations. a. Exceptionally keen and alert. b. Usually quick to understand and to learn. c. Grasps instructions with average ability. d. Slow to grasp instructions. LITERATURE CITED Books Cronbach, L. J. and Closer, Goldine C. Psychological Tests and Personnel Decisions (Urbana: University of Illinois Press, 1956). Ells, R. and Walton, C. Conceptual Foundations of Business. (Homewood, Illinois: Irwin Press, 1961). Guion, Robert M. Personnel Testing (New York: McGraw Hill Book Company, 1965). Hull, C. L. Aptitude Testing (New York: Harcourt, Brace and World, Inc., 1928). Kellogg, Marion S. Closing the Performance Gap (New York: American Management Association, 1967). Lawshe, C. H. Principles of Personnel Testing (New York: McGraw Hill Book Company, 1948). Lipsett, Lawrence. Personnel Selection and Recruitment (New York: Prentice-Hall, Inc., 1964). Lopez, Felix M. , Jr. Personnel Interviewing: Theory and Practice (New York: McGraw Hill Book Company, 1965). Articles Ash, Phillip, "Selection Techniques and the Law," Personnel, XLIV (November-December, 1967), 8-17. Jurgensen, C. E., "Report on the Classification Inventory: A Personality Test for Industrial Use," Journal of Applied Psychology, XXVIII, 445-460. Katzell, R. A., "Industrial Psychology," Annual Review of Psychology, III, 237-268. Kirkwood, John H., "To Test or Not to Test?", Personnel, XLIV (November-December, 1967), 18-26. Stromberg, E. L., "Testing Programs Draw Better Applicants," Personnel Psychology, I, 21-29. Uhrbroch, R. S., "Mental Alertness Tests as Aids in Selecting Employees," Personnel, XII (November-December, 1935), 229- 237. 63 63 Miscellaneous Equal Employment Opportunity Commission, Guidelines on Employment Testing Procedures (Washington, D. C.: Equal Employment Opportunity Commission, 1966). Jemssen, Ward J., National Cash Register Test Battery Instruc¬ tion Manual for Prospective Check-Out Cashiers (Dayton, Ohio; The National Cash Register Company, 1961). McMurry, Robert N. and Arnold, Phyllis D., Science Research Association Pictorial Reasoning Test Preliminary Examiner's Manual (Chicago; Science Research Associates, Inc., 1966).